The present invention relates generally to rigid disk substrates for use in magnetic recording applications. More specifically, the present invention relates to a substrate formed of a ceramic-metal matrix composite. The ceramic-metal matrix composite may be clad with metal to enhance the smoothness of the rigid disk substrate.
Personal computers have become common tools for data manipulation and data storage. Computers store and retrieve data using magnetic recording technology, where a magnetic film supported on a rigid disk substrate is used as the recording medium.
Magnetic recording involves use of a head, which includes a ring of magnetic material having a very narrow gap and a wire wrapped around the ring. To record or write information, a current is passed through the wire to generate a very intense field in the vicinity of the gap. If a recording medium is passed very close to the gap, it becomes permanently magnetized in response to the current. The recording is retrieved or played back by moving the recorded medium past the head and recovering the small voltage induced in the wire caused by a change in flux through the ring as the recorded magnetic information passes by the gap.
The storage capacity of typical magnetic recording disks used in personal computer hard disk drives has grown by over 70% per year from 1980 to 1991, and continues to have a high growth rate today. The density at which information can be recorded (written) and reproduced (read) on a disk surface is determined by the dimensions of the head and the accuracy with which the head can be positioned, by the magnetization level that can be achieved in the recording medium, and by the accuracy and sophistication of the reproduction electronics.
Data are recorded on the surface of a disk spinning typically at rotation speeds of approximately 5400 revolutions per minute (rpm). To avoid contact and wear, the head is designed as a slider that glides or flies just above the disk surface. Very high linear track densities can be achieved by reducing the separation between the head surface and the recording medium, thus increasing the storage capacity of the disk.
Therefore, to maximize the resolution in writing and reading and hence maximize the storage capacity of a disk, it is necessary to have a very small head with a very small gap and a very narrow track width within which information is recorded. Most importantly, it is necessary to have an extremely small flying height or separation between the head surface and the recording medium. Both the field intensity during writing and the sensitivity during reading drop dramatically as the flying height increases beyond about one-third the length of the smallest piece of recorded information. In addition, future generations of recording heads may actually be designed to contact the disks during operation. Therefore, surface roughness of the disks is extremely critical, and surfaces must be optically smooth because the bit lengths are of the same order of magnitude as the wavelength of light. Further, disk deflection during rotation, described hereinbelow, must be eliminated.
As shown in FIG. 1, a typical magnetic recording disk 1 used today is comprised of a rigid disk substrate 3 made of an aluminum-magnesium (Al--Mg) alloy, a nickel-phosphorous (Ni--P) layer 5 covering the Al--Mg substrate, a magnetic recording layer 7 covering the Ni--P layer, and a carbon-based protective layer 9 covering the magnetic recording layer. The Ni--P layer 5 serves as a hard coating that protects the Al--Mg substrate and also is generally used for texturing the surface of the substrate 3 before the magnetic recording layer 7 is deposited. Texturing produces a grooved surface which assists and improves head aerodynamics.
A current trend in disk drive technology is to optimize rigid disk substrates by making them thinner, more rigid, and harder so that more disks can be stacked within a given space and also so that they can withstand handling when used in portable disk drives. Preferably, the substrates would have a high specific modulus of elasticity E/.rho., where E is the modulus of elasticity and .rho. is the density of the substrate material, at a substrate thickness that is less than 1 mm. A typical thickness of a conventional Al--Mg substrate ranges roughly from 0.6 to 0.9 mm. Hardness is important because a recording head is subject to head "slaps" or contacts with a spinning disk. These slaps can have a force of 500 to 1000 G. Head slaps are particularly prevalent in conventional Al--Mg substrates due to air turbulence-induced deflections. Because of these deflections, a conventional substrate will not have a flat profile while spinning, but will have a slight wobble causing the substrate to slap the head if the amount of wobble exceeds the flying height of the head.
Other considerations in choosing an alternative and improved rigid disk substrate material is its compatibility with standard deposition processes used to deposit the various layers in a magnetic recording disk, including the Ni--P layer, the magnetic recording layer, and the protective coating layer.
In view of the above-mentioned problems and considerations, the present invention contemplates a ceramic-metal matrix composite for use as a rigid disk substrate. The ceramic-metal matrix composite is comprised of a metal matrix material to which is added a ceramic material to improve mechanical properties such as stiffness and hardness of the metal matrix material, for example. The surface of the ceramic-metal matrix composite may be clad with a metal layer for minimizing topological irregularities in the ceramic-metal matrix composite in order to enhance the smoothness of the surface for subsequent formation of a magnetic recording layer and to make the surface of the ceramic-metal matrix composite compatible with subsequent manufacturing processes. The ceramic-metal matrix composite of the present invention is stronger, stiffer, and exhibits other significant improvements over materials used in conventional Al--Mg rigid disk substrates at a comparable cost. There are several alternative processes for producing the clad structure of the rigid disk substrate.
One such ceramic-metal matrix composite material for the use contemplated by the present invention is described in U.S. Pat. No. 5,486,223, which is incorporated herein by reference.
In recent years ceramic-metal matrix composites have become desirable materials because of improvements in stiffness, strength, and wear properties. Basic ceramic-metal matrix composites are made typically with aluminum, titanium, magnesium, or alloys thereof as the metal matrix material. A selected percentage of ceramic material, within a specific range, is added to the metal matrix material to form the composite. Typical ceramic additives include boron carbide, silicon carbide, titanium diboride, titanium carbide, aluminum oxide, and silicon nitride.
Most known ceramic-metal matrix composites are made by a conventional process that introduces the ceramic material into a molten metal matrix. In order for the improved properties to be realized, the molten metal generally must wet the ceramic material so that segregation or clumping of the ceramic material is minimized. Numerous schemes with varying degrees of success have been utilized to improve the dispersion of the ceramic material in the molten metal.
Recently, powder metallurgy consolidation has emerged as an attractive alternative method for fabricating metal matrix composites, where the powders are compacted by means of hot pressing and vacuum sintering to achieve a high density billet. By following certain pressing and sintering techniques, a billet of 99% theoretical density can be achieved.
One problem encountered in composites of aluminum and certain ceramic materials is the thermodynamic instability of those ceramic materials in molten aluminum heated to excessively high temperatures. This instability leads to the formation of unwanted precipitates at grain boundary interfaces, which are believed to have detrimental effects on the mechanical properties of the resulting composite.